Telegram Channel Scraper

Scrape public Telegram channels with no login and no API key, using Telegram's own server-rendered web preview at https://t.me/s/<channel>. No bot token, no phone number, no MTProto.

What it does

• Accepts a list of channels in any common form: durov, @durov, or https://t.me/durov.
• Reads the channel header: title, description, and counters (subscribers, photos, videos, links).
• Returns one row per post with text (links and hashtags preserved), date, views, author, media info, and any forwarded/source link.
• Pages back through older posts automatically via ?before=<id> until it hits maxItems or the channel's earliest post.
• Dedupes posts by message id.

Only works for public channels that expose a web preview. Private channels, groups, and channels with the preview disabled will return a NOT_FOUND diagnostic row.

Input

Field	Type	Default	Description
channels	string[]	["telegram","durov"]	Channels to scrape (username, @handle, or t.me link).
maxItems	integer	100	Max posts per channel.
includeChannelInfoRow	boolean	true	Emit one channel summary row before each channel's posts.
proxyConfiguration	object	off	Optional. The t.me/s/ preview is a public, no-auth endpoint with no anti-bot, so a proxy is not required and is off by default. Only enable it (Apify Proxy, datacenter is fine) if you hit IP rate limits on high-volume runs.

Output

One channel-info row per channel (when enabled), then one row per post:

{
  "ok": true,
  "rowType": "post",
  "channel": "durov",
  "messageId": 505,
  "postUrl": "https://t.me/durov/505",
  "text": "TON is #1 in annual staking rewards…",
  "date": "2026-05-08T17:41:53.000Z",
  "views": "3.76M",
  "author": "Pavel Durov",
  "hasMedia": true,
  "mediaType": "photo",
  "mediaUrl": "https://cdn4.telesco.pe/file/…",
  "link": null,
  "channelTitle": "Pavel Durov",
  "subscribers": "11M"
}

mediaType is one of photo, video, roundvideo, voice, sticker, document, or null.

Nullable fields: the preview only exposes what Telegram renders, so several fields are commonly null: text (media-only posts), views (small/old channels), author (most channels don't sign posts), mediaType / mediaUrl (text-only posts), and link (posts that aren't forwards and have no link preview). On the channel-info row, subscribers, photosCount, videosCount, and linksCount can be null when the header doesn't show that counter.

Billing

Charged per post row (post event). Channel-info rows and diagnostic rows (BAD_INPUT, NOT_FOUND, NO_RESULTS, network errors) are not charged.

Troubleshooting

• NOT_FOUND / NO_RESULTS — the channel is private, deleted, a group, or has its web preview disabled. Only public channels with a working t.me/s/<name> page can be scraped.
• BAD_INPUT — no usable channel was provided. Pass a username, @handle, or t.me link.
• RATE_LIMITED on large runs — lower maxItems or enable the optional proxy to spread requests across IPs.